A Study on Daily Peak Electric Load Forecasting Using Radial Basis Function Netwrok

The electric load forecasting is a challenging problem because variations of load might occur in an hour or a day period. These variations depend on many variables such as weather conditions, the day of the week, time of day, seasonal factors, social activities, and social economic factors. In this paper, we propose the load forecasting method based on the RBF neural network, which is an efficient technique for the daily peak load forecasting in Korea. The 2014-2016 dataset in Korea was employed to evaluate and validate the proposed approach. Obtained results were compared with the feed-forward network. The study showed that the proposed approach can be applied to forecast the daily peak electric load effectively.